Prostate cancer is the most commonly diagnosed cancer and increasing with the expanding aging population. Epithelial transformation is predicated on its phenotypic, epigenetic, and genomic changes. The reciprocating phenotypic changes in the adjacent stromal microenvironment, or stromal co-evolution, are supportive of the adjacent epithelia by the differential production of extracellular matrix and growth factors. However, the classic demonstration of isolated carcinoma associated fibroblasts, grown ex vivo for a period of time, to subsequently induce tumorigenesis on non-tumorigenic epithelia in vivo is suggestive of epigenetic imprinting. We have characterized the PCa associated stroma in mouse models and human disease to have TGF-?-dependent epigenetic changes not observed in stromal fibroblasts associated with benign prostate epithelia. Two independent conditional stromal knockout mouse models of the TGF-? type II receptor (Tgfbr2) models develop prostate adenocarcinoma that support castrate resistance. In the last funding cycle we also identified that the knockout of Tgfbr2 up regulates Stat3 signaling. We find that there is elevated epigenetic silencing of DNA damage repair genes associated with a disproportionate amount of DNA double stranded breaks in the stromal fibroblastic cells. We will test the hypothesis that, TGF-? signaling regulation of tumor suppressors in stromal fibroblasts mediates prostate cancer progression. The mechanism of the epigenetic changes in the tumor microenvironment and its role in prostate cancer progression is the focus of Aim 1. It is still poorly understood how in prostate cancer androgen receptor function is abnormally switched from a pro-differentiation role in development and normal adult prostate homeostasis, to one pro-proliferation in cancer. Since prostate epithelial androgen responsiveness is mediated through the stromal androgen signaling, the role of stromal epigenetic signaling will be a focus of Aim 2. The role of Stat3 in tumor progression and stromal androgen signaling will be determined in terms of its role on DNA methyl transferase and androgen receptor activity. The clinical relevance of the co-evolving stromal changes observed we be tested in a retrospective case control trial for patient outcome following radiation therapy for localized prostate cancer. We are uniquely positioned to address the role of cancer associated stromal changes that occurs in tumorigenesis and castrate resistant growth, based on TGF-?'s role in tumor suppressor silencing and defining stromal heterogeneity through cell culture, tissue recombination xenograft, and transgenic mouse models. .